This proposed continuing Program Project will use prospectively collected dietary data, frozen plasma and DNA, and tumor tissue to address hypotheses regarding cancers of the prostate, colon and rectum, breast, and brain (glioma). This Program supports, and depends on, the continued follow-up of 51,529 men who have completed extensive dietary questionnaires every four years since 1986 (the Health Professionals Follow-up Study, HPFS), and is also closely linked to the Nurses' Health Study (NHS) of 121,700 women. The Program Project is a founding member of, and contributes data and DNA samples to; the NCI sponsored Cohort Consortium on genetic determinants of cancer risk. The proposed continuation will extend and refine observations from the first 19 years of follow-up and will also address new hypotheses related to both cancer incidence and survival. Cross-cutting themes are the investigation of vitamin D, inflammatory, and energy balance pathways as they relate to cancer incidence. Particular emphasis is given to potentially modifiable determinants of cancer risk and survival. Project 1 examines hypotheses relating vitamin D intakes, blood levels, and metabolizing genes and consumption of dairy products to both localized and fatal prostate cancer. Use of statins, aspirin, and NSAID's and of insulinemia (weight and weight gain, dietary insulin index, plasma C-peptide, and adiponectin) will also be examined detail. Specific dietary factors will be addressed in relation to survival among men with prostate cancer. Project 2 examines vitamin D blood levels, intakes, and metabolizing genes in relation to the diagnosis of colorectal cancer and adenomas, including associations with histological and molecular characteristics of tumors. The use of anti-inflammatory agents and their interactions with polymorphisms in related genes, and of insulin, adiponectin, and IFG-1 pathways to risk of risk of colon cancer will also be evaluated. Specific dietary factors, weight change, physical activity, and use of anti-inflammatory agents will be examined in relation to survival among patients with colorectal cancer. Project 3 pools data from 21 major prospective studies of diet and cancer. Precise and unique information has already been obtained for breast, lung, colon, kidney, ovarian, and pancreatic cancer, and the proposed work will extend analyses to prostate cancer, estrogen receptor negative breast cancer, and gliomas. Project 4 addresses methodological issues in epidemiological studies of genetic polymorphisms, diet and risk of cancer. These include the analysis of multiple polymorphisms within metabolic pathways and the effects of, and adjustment for, measurement errors in both diet and genotypes. Lav summary: These highly interrelated projects integrate dietary factors, related non-dietary factors, blood nutrient and hormone levels, genetic susceptibility, and molecular characteristics of tumors. The findings will contribute importantly to the prevention and prognosis of major cancers of men and women.